Project Summary/Abstract Heart failure with preserved ejection fraction (HFpEF) comprises half of all HF, has high morbidity and is growing in prevalence. Traditional HF therapy does not improve outcomes in HFpEF, potentially owing to heterogeneous definitions of HFpEF itself. Societal and clinical trial definitions of HFpEF lack consensus, relying largely on resting cardio-centric measures (e.g., hypertrophy, diastolic filling, filling pressure) and natriuretic peptide levels. Furthermore, the cardinal manifestation of HFpEF is exertional intolerance (with or without overt congestion), the etiology of which is frequently not captured by resting characterization. Our group has used comprehensive cardiopulmonary exercise testing (CPET) as a quantitative probe of global metabolic capacity (peak VO2) alongside measures of multi-organ reserve in HF. Through simultaneous quantitation of invasive hemodynamics, blood gases, cardiac function, arterial tonometry and gas exchange patterns during exercise in individuals with conventionally defined HFpEF, we have started to delineate contributions of impaired cardiac, pulmonary, vascular, and peripheral musculoskeletal reserve capacity that are not evident at rest. We further hypothesized that distinct metabolic defects underlie these findings, identifying selected circulating metabolites associated with HF-defining phenotypes in humans and animal models. While these preliminary studies suggest that mapping metabolic responses during exercise may resolve phenotypic heterogeneity within HFpEF, studies addressing this approach in large populations with well-characterized phenotypes during exercise are lacking. Here, we address this gap by characterizing suspected HFpEF via measures of sympathetic nervous system, cardiac, vascular, and musculoskeletal metabolic function during exercise in 1312 individuals via CPET and metabolite profiling. We hypothesize that exercise will unmask predominant organ-specific reserve deficits representing distinct HFpEF ?pathophenotypes.? We further hypothesize that metabolic patterns associated with these pathophenotypes will be dysregulated early in HFpEF progression, identifying targetable pathways central to HFpEF. In Aim 1, we identify predominant organ-specific pathophenotypes in 1312 individuals with suspected HFpEF in a prospective cohort study at our center (MGH-ExS study). In Aim 2, we identify metabolic correlates of HFpEF pathophenotypes via targeted metabolite profiling in MGH-ExS and evaluate these metabolite-pathophenotype associations in the community (Framingham Heart Study [FHS] 3rd Generation). In Aim 3, we test association of metabolite- and CPET-based HFpEF pathophenotypes with long-term HF in the MGH-ExS and in the community (Health ABC study; FHS). Our team has extensive experience in exercise physiology, HF, metabolite profiling, and bioinformatics uniquely suited to this application. Successful completion will enhance precision-definitions of HFpEF and will provide a unique resource (CPET and metabolite data) for the scientific community.